Ebook Blood pressure monitoring in cardiovascular medicine and therapeutics: Part 2

(BQ) Part 2 book Blood pressure monitoring in cardiovascular medicine and therapeutics presents the following contents: Importance of heart rate in determining cardiovascular risk; sodium, potassium the sympathetic nervous system and the renin–angiotensin system - impact on the circadian variability in blood pressure; prognostic value of ambulatory blood pressure monitoring; circadian rhythm of myocardial infarction and sudden cardiac death;...

From: Contemporary Cardiology:

Blood Pressure Monitoring in Cardiovascular Medicine and Therapeutics

Edited by: W B White © Humana Press Inc., Totowa, NJ

Importance of Heart Rate

in Determining Cardiovascular Risk

A body of evidence indicates that subjects with tachycardia are more likely

to develop hypertension (1–3) and atherosclerosis in future years (4–6)

How-ever, the connection between heart rate and the cardiovascular risk has long beenneglected, on the grounds that tachycardia is often associated with the traditionalrisk factors for atherosclerosis, such as hypertension or metabolic abnormalities

(7) A high heart rate is currently considered only an epiphenomenon of a

com-plex clinical condition rather than an independent risk factor However, mostepidemiogic studies showed that the predictive power of a fast heart rate for car-diovascular disease remains significant even when its relative risk is adjusted for

all major risk factors for atherosclerosis and other confounders (4–7) In this

chapter, the results of the main studies that dealt with the relation between cardia and cardiovascular morbidity and mortality will be summarized, and thepathogenesis of the connection between fast heart rate and cardiovascular dis-ease will be the focus

tachy-EPIDEMIOLOGIC EVIDENCE

The heart rate was found to be a predictor for future development of

hyper-tension as far back as in 1945 (8) This finding was subsequently confirmed by

the Framingham study, in which the predictive power of the heart rate for future

development of hypertension was similar to that of obesity (3) Several other more recent reports have confirmed those findings (1,2,9) The heart rate was found to be also a predictor of myocardial infarction (10,11) and of cardiovas- cular morbidity in general (5,8) A body of evidence indicates that tachycardia

is also related to increased risk of cardiovascular mortality This association was

shown by Levy et al in a survey of over 20,000 Army officers (8) Thereafter,

a number of other studies confirmed this finding, showing that the resting heartrate was a powerful predictor of death from cardiovascular and noncardiovascular

causes (4,5,6,12–15) The data related to sudden death were particularly

impres-sive, especially in the Framingham study, in which a sharp upward trend in

mor-tality was found in the men divided by quintiles of heart rate (6) Also, in the

Chicago studies a strong association was found between heart rate and suddendeath, but the relation was U-shaped, because of an excess of mortality also in

the subjects with very low heart rates (4).

The relationship between heart rate and cardiovascular mortality persists into

old age This was shown by the Framingham (6,16) and the NHANES (5) studies

performed in general populations and by two more recent studies conducted in

elderly subjects (13,14) In the CASTEL study (13), the predictive power of heart

rate for mortality was 1.38 for the men with a heart rate > 80 beats/minute (bpm)(top quintile) compared to those of the three intermediate quintiles, and 0.82 forthe men with a heart rate < 60 bpm (bottom quintile) The relation between heartrate and mortality was particularly strong for sudden death, with an adjustedrelative risk of 2.45 for the subjects in the top quintile as compared to those inthe three intermediate quintiles In the CASTEL study, no significant associationbetween heart rate and mortality was found in the women In another study per-

formed on elderly men and women combined (14), a 1.14 times higher

probabil-ity of developing fatal or nonfatal myocardial infarction or sudden death wasfound for an increment of 5 bpm of heart rate recorded over the 24 h

In the Framingham study, the relationship of heart rate with morbidity and

mortality was analyzed also within hypertensive individuals (15) followed up for

36 yr For a heart rate increment of 40 bpm the age-adjusted and pressure-adjusted relative risk for cardiovascular mortality was 1.68 in males and1.70 in females For sudden death, the adjusted odds ratios were 1.93 and 1.37,respectively These relationships were still significant after adjusting for smok-ing, total cholesterol, and left ventricular hypertrophy

systolic-blood-The heart rate was found to be a strong predictor of cardiovascular mortalityalso in patients with myocardial infarction This association was found in the

Norwegian Timolol Multicenter Study (17) and in a study by Hjalmarson et al.

(18) in which the total mortality was 14% in the subjects with an admission heart

rate < 60 bpm, 41% in the subjects with a heart rate > 90 bpm, and 48% in thosewith a heart rate >110 bpm In a subsequent study, Disegni et al found a doubledmortality risk in postmyocardial infarction patients with a heart rate > 90 bpm

compared to subjects with a heart rate < 70 bpm (19) Two analyses performed

in larger datasets confirmed the results of the above studies In the GUSTO study

(20), a high heart rate emerged as a potent precursor of mortality, and in the

GISSI-2 trial (21), the predischarge heart rate was a stronger predictor of death

than standard indices of risk, such as left ventricular dysfunction or ventriculararrhythmias It is noteworthy to observe that tachycardia in postmyocardial infarc-tion patients cannot be considered simply as a marker of heart failure, as its pre-dictive power appeared more evident in the subjects with no or mild signs of

congestive heart failure (18,19) In a recent study, we found that the predictive

power of heart rate for mortality in subjects with acute myocardial infarctionremained significant also after adjusting for numerous confounders, includingclinical and echocardiographic signs of left ventricular dysfunction (Palatini etal., unpublished observations) (Fig 1)

PATHOGENETIC CONSIDERATIONS

The pathogenetic connection between fast heart rate and cardiovascular riskcan be explained according to several different mechanisms (Fig 2) The heart

Fig 1 Relative risks (RR) and 95% confidence limits (CL) for 1-yr mortality in 250 men

divided according to whether their heart rate was < 80 bpm or  80 bpm on the seventh day after admission to the hospital for acute myocardial infarction Unadj = unadjusted relative risk; age-adj = relative risk adjusted for age; risk-adj = relative risks adjusted for age, CK-

MB peak, echocardiographic left ventricular ejection fraction, diabetes, history of sion, current smoking, history of angina, Killip class, thrombolysis and `-blocker therapy;

hyperten-p-values relate to the results of Cox regression analyses.

rate can be considered as a marker of an underlying clinical condition related

to the risk or a consequence of a latent chronic disease However, experimentalevidence suggests that a high heart rate should be regarded as a pathogeneticfactor in the induction of the risk as well In fact, tachycardia favors the occur-

rence of atherosclerotic lesions by increasing the arterial wall stress (22) and impairs arterial compliance and distensibility (23) Moreover, the mean blood

Fig 2 Mechanisms of the connection between heart rate and cardiovascular morbidity and

mortality The heart rate can be a marker of risk or a consequence of an underlying disease, but can exert a direct action in the induction of the risk as well LV = left ventricular; BP

= blood pressure, w = increased, ¦ = decreased.

pressure has been found to be higher in subjects with faster heart rate (24) This

can be explained by the increase in the total time spent on systole because of theshortening of diastolic time

The experimental evidence for a direct role of tachycardia in the induction ofarterial atherosclerotic lesions was provided by studies performed in cynomol-gus monkeys Beere et al were the first to demonstrate that reduction of heart rate

by ablation of the sinoatrial node could retard the development of coronary lesions

in these animals (25).

Bassiouny et al studied the effect of the product of mean heart rate and mean

blood pressure (so-called hemodynamic stress) on the aorta of the monkeys (26)

and found a striking positive relationship between the hemodynamic stress indexand maximum atherosclerotic lesion thickness Similar results were obtained byKaplan et al., who found a significant relationship between naturally occurring

differences in heart rate and atherosclerotic coronary lesions in monkeys (27).

As mentioned earlier, heart rate can be considered as a marker of an abnormalclinical condition This is suggested by the relationship found in several studies

between heart rate and many risk factors for atherosclerosis (28–30) In four

different populations studied in the Ann Arbor laboratory, we found that theheart rate was correlated with blood pressure, degree of obesity, cholesterol, tri-

glycerides, postload glucose, and fasting insulin (Table 1) (31,32) In other words,

subjects with a fast heart rate exhibited the features of the insulin-resistancesyndrome If one assumes that a fast heart rate is the marker of an abnormal auto-

nomic control of the circulation, as demonstrated by Julius et al (33,34), it is easy

to understand why subjects with tachycardia develop atherosclerosis and vascular events In fact, several studies performed in the Ann Arbor and otherlaboratories indicate that sympathetic overactivity can cause insulin resistance

cardio-(Fig 3) This can be obtained through acute (35) as well as chronic (36)

stimu-lation of `-adrenergic receptors It has been shown that chronic stimustimu-lation of

Table 1 Correlation Coefficients Between Resting Heart Rate and Other Clinical Variables in Three General and One Hypertensive Populations

Population SBP DBP BMI CT TG GL INS

Data from ref 7.

`-receptors causes the conversion from a small to a larger proportion of

insulin-resistant fast-twitch muscles (36) An insulin-resistance state can be obtained

also through a vasoconstriction mediated by _-adrenergic receptors, as shown

by Jamerson et al in the human forearm (37) Conversely, the _-adrenergic blockade can improve insulin sensitivity in patients with hypertension (38).

The connection between high heart rate and mortality can be explained also

by an unrecognized underlying disease, and tachycardia can reflect poor

physi-cal fitness or loss of cardiac reserve (4,6,13) In fact, an impaired left

ventricu-lar contractility may be an early clinical finding in asymptomatic hypertensive

individuals, as demonstrated in the Padova (39) and Ann Arbor (40) laboratories.

To rule out this possibility, in some studies the subjects who died within the first

years after the baseline evaluation were eliminated (6,13,16) However, in all of

those studies, the heart rate–mortality association remained significant, ing that tachycardia was not only a marker of latent left ventricular failure or ofloss of vigor

indicat-Fig 3 Pathogenesis of the connection between tachycardia and insulin resistance

Tachy-cardia is a marker of the underlying sympathetic overactivity SNS = sympathetic nervous system, w = increased, ¦ = decreased.

Besides causing the development of atherosclerotic lesions, a fast heart ratecan also favor the occurrence of cardiovascular events, as shown by the Framing-

ham study (6,12,16) The relationship appeared weak for nonfatal cardiovascular

events but was strong for fatal cardiovascular events Moreover, as mentioned

earlier, tachycardia can facilitate sudden death (4,6,13) The reasons for this

connection can be of a different nature Sympathetic overactivity underlying afast heart rate can facilitate the occurrence of coronary thrombosis through plate-

let activation and increased blood viscosity (31) Subjects with tachycardia are

more prone to ventricular arrhythmias It is known that a heightened sympathetic

tone can promote the development of left ventricular hypertrophy (41), which facilitates the occurrence of arrhythmias (42) Moreover, tachycardia increases oxygen consumption and ventricular vulnerability (7,43) The latter mecha-

nisms are important chiefly in subjects with acute myocardial infarction

LOOKING FOR A THRESHOLD VALUE

The current definition of tachycardia is a heart rate > 100 bpm Recent resultsobtained in our laboratory with mixture analysis suggest that this value is prob-ably too high In fact, in three general and one hypertensive populations, wefound that the distribution of heart rate was explained by the mixture of twohomogeneous subpopulations, a larger one with a “normal” heart rate and asmaller one with a “high” heart rate The partition value between the two sub-populations was around 80–85 bpm Furthermore, in almost all of the epidemio-logic studies that showed an association between heart rate and death fromcardiovascular or noncardiovascular causes, the heart-rate value above which

a significant increase in risk was seen was below the 100-bpm threshold (44)

(Table 2) On the basis of the above data we suggested that the upper normal value

of heart rate should be set at 85 bpm (44).

THERAPEUTIC CONSIDERATIONS

Although there is no doubt that a fast heart rate is independently related tocardiovascular and total mortality, it is not known whether the reduction of heartrate can be beneficial in prolonging life No clinical trial has been implemented

as yet in human beings with the specific purpose of studying the effect of cardiacslowing on morbidity and mortality This issue was dealt with by Coburn et al

in mice by studying the effect of digoxin administration (45) Survival increased

by 29% in the digoxin-treated males and by 14% in the treated females, in parison with two groups of untreated mice (control groups), indicating that aheart-rate reduction may confer an advantage in terms of longevity

com-A beneficial effect of heart-rate reduction in retarding the development ofatherosclerotic lesions was demonstrated by Kaplan et al with `-blocker admin-

istration in cynomolgus monkeys (46) After 26 mo of propranolol treatment, the

socially dominant animals showed a reduced development of coronary arterylesions in comparison to a group of untreated monkeys of the control group Thissuggests that heart-rate reduction with `-blockers is beneficial in preventingatherosclerotic lesions, but only in animals exposed to a high environmental stress.Most of the information on the effect of `-blockers on heart rate and morbidityand mortality in human beings comes from results obtained in post-myocardial-infarction patients The reduction in heart rate obtained varied greatly among thetrials, from 10.5% to 22.8% `-Blocking treatment appeared beneficial in thosepatients in whom the heart rate was reduced by 14 bpm or more, whereas for a

heart-rate reduction <8 bpm, no benefit was apparent (47) Moreover, the

advan-tage of treatment was virtually confined to patients with a heart rate of >55 bpm

In 26 large, placebo-controlled trials with a long-term follow-up, `-blockersproved effective primarily in reducing sudden death and death resulting from

pump failure (47–51) An almost linear relationship was found between tion in resting heart rate and decreased mortality (48,52).`-Blockers with intrin-sic sympathomimetic activity, such as pindolol or practolol, showed only littleeffect on mortality

reduc-Similar beneficial effects were obtained in patients with congestive heart

fail-ure (53) Carvedilol caused a marked reduction in mortality in subjects with gestive heart failure (54), but only in patients with a high heart rate (>82 bpm).

con-Table 2 Heart Rate Threshold Values Above Which a Significant

Increase in Mortality Was Found in Eight Epidemiologic Studies

HR threshold value

Reference Men Women Results of the study

Levy et al., 1945 (8) 99 — Increased 5-yr cardiovascular mortality

in men

Dyer et al., 1980 (4) 79 — Increased 15-yr all-cause mortality in the

men of the People Gas Co study

Dyer et al., 1980 (4) 86 — Increased 5-yr all-cause mortality in the

men of the Heart Association study

Dyer et al., 1980 (4) 89 — Increased 17-yr all-cause mortality in the

men of the Western Electric study

Kannel et al., 1985 (6) 87 87 Increased 26-yr sudden death mortality

rate in men

Gillum et al., 1991 (5) 84 84 Increased 10-yr all-cause mortality in

black and white men and in black women

Gillman et al., 1993 (15) 84 84 Increased 36-yr all-cause mortality in

hypertensive men and women

Palatini et al., 1999 (13) 80 84 Increased 12-yr cardiovascular mortality

in elderly men

HR = heart rate in bpm.

The results obtained in hypertensive subjects (55) were less impressive,

prob-ably the result of the untoward effects of `-blockers on high-density lipoprotein

(HDL) cholesterol and triglycerides (56) However, the effect of `-blockers in

hypertensive patients was never examined in relation to the subjects’ heart rates

at baseline

If the unsatisfactory effects of `-blockers in hypertension are the result oftheir unfavorable effects on plasma lipids, the use of drugs which reduce bloodpressure and heart rate without altering the lipid profile appears warranted Non-

dihydropyridine-calcium antagonists (57,58) have been shown to be neutral on

the metabolic profile and could, thus, be more effective in preventing cular mortality in hypertensive subjects with tachycardia In addition to having

cardiovas-a periphercardiovas-al cardiovas-action, some of them ccardiovas-an cross the blood-brcardiovas-ain bcardiovas-arrier cardiovas-and decrecardiovas-ase

sympathetic outflow (58).

Diltiazem and verapamil have been shown to be effective in reducing the risk

of cardiac events (59–61), but their depressive action on cardiac inotropism

makes them unsuitable for patients with acute myocardial infarction and severeleft ventricular dysfunction The new long-acting calcium antagonists that selec-

tively block voltage-dependent T-type calcium channels (62,63) reduce heart rate

without manifesting a depressant effect on myocardial contractility and could,

thus, be indicated also for subjects with congestive heart failure (64).

Centrally active antihypertensive drugs that decrease heart rate through tion of the sympathetic discharge from the central nervous system should have

reduc-a good potentireduc-al for the trereduc-atment of the hypertensive preduc-atient with freduc-ast hereduc-art rreduc-ate.Unfortunately, the use of clonidine, _-methyldopa, guanfacine, and guanabenz

is limited by the frequent occurrence of side effects, like dry mouth, sedation, and

impotence (65) Moxonidine and rilmenidine are new antihypertensive agents

acting on the I1-imidazoline receptors of the rostro-ventrolateral medulla of thebrainstem and do not have most of the side effects encountered with the centrally

acting agents (65,66) Moreover, these drugs proved effective in improving the metabolic profile in the experimental animal (67) and also in human studies (68).

The goal of antihypertensive treatment should be not only to lower the bloodpressure but also to reverse those functional abnormalities that often accompanythe hypertensive condition Therefore, a therapy that not only reduces bloodpressure effectively but also decreases the heart rate and improves metabolicabnormalities should be sought

REFERENCES

1 Selby JV, Friedman GD, Quesenberry CP Jr Precursors of essential hypertension: pulmonary function, heart rate, uric acid, serum cholesterol, and other serum chemistries Am J Epidemiol 1990;131:1017–1027.

2 Reed D, McGee D, Yano K Biological and social correlates of blood pressure among Japanese men in Hawaii Hypertension 1982;4:406–414.

3 Garrison RJ, Kannel WB, Stokes J III Incidence and precursors of hypertension in young adults Prev Med 1987;16:235–251.

4 Dyer AR, Persky V, Stamler J, et al Heart rate as a prognostic factor for coronary heart ease and mortality: findings in three Chicago epidemiologic studies Am J Epidemiol 1980;112: 736–749.

dis-5 Gillum RF, Makuc DM, Feldman JJ Pulse rate, coronary heart disease, and death: the NHANES

I epidemiologic follow-up study Am Heart J 1991;121:172–177.

6 Kannel WB, Wilson P, Blair SN Epidemiologic assessment of the role of physical activity and fitness in development of cardiovascular disease Am Heart J 1985;109:876–885.

7 Palatini P, Julius S Heart rate and the cardiovascular risk J Hypertens 1997;15:3–17.

8 Levy RL, White PD, Stroud WD, et al Transient tachycardia: Prognostic significance alone and in association with transient hypertension JAMA 1945;129:585–588.

9 Paffenbarger RS Jr, Thorne MC, Wing AL Chronic disease in former college students—VIII Characteristics in youth predisposing to hypertension in later years Am J Epidemiol 1968;88: 25–32.

10 Schroll M, Hagerup LM Risk factors of myocardial infarction and death in men aged 50 at entry A ten-year prospective study from the Glostrup population studies Dan Med Bull 1977; 24:252–255.

11 Medalie JH, Kahn HA, Neufeld HN, et al Five-year myocardial infarction incidence—II Association of single variables to age and birthplace J Chronic Dis 1973;26:329–349.

12 Goldberg RJ, Larson M, Levy D Factors associated with survival to 75 years of age in aged men and women The Framingham study Arch Intern Med 1996;156:505–509.

middle-13 Palatini P, Casiglia E, Julius S, et al Heart rate: a risk factor for cardiovascular mortality in elderly men Arch Int Med 1999;159:585–592.

14 Aronow WS, Ahn C, Mercando AD, et al Association of average heart rate on 24-hour ambulatory electrocardiograms with incidence of new coronary events at 48-month follow-

up in 1,311 patients (mean age 81 years) with heart disease and sinus rhythm Am J Cardiol 1996;78:1175–1176.

15 Gillman MW, Kannel WB, Belanger A, et al Influence of heart rate on mortality among persons with hypertension: The Framingham Study Am Heart J 1993;125:1148–1154.

16 Kannel WB, Kannel C, Paffenbarger RS Jr, et al Heart rate and cardiovascular mortality: The Framingham Study Am Heart J 1987;113:1489–1494.

17 The Norwegian Multicenter Study Group Timolol-induced reduction in mortality and farction in patients surviving acute myocardial infarction N Engl J Med 1981;304:801–807.

rein-18 Hjalmarson A, Gilpin EA, Kjekshus J, et al Influence of heart rate on mortality after acute myocardial infarction Am J Cardiol 1990;65:547–553.

19 Disegni E, Goldbourt U, Reicher-Reiss H, et al The predictive value of admission heart rate on mortality in patients with acute myocardial infarction J Clin Epidemiol 1995;48: 1197–1205.

20 Lee KL, Woodlief LH, Topol EJ, et al Predictors of 30-day mortality in the era of reperfusion for acute myocardial infarction Results from an international trial of 41,021 patients Circula- tion 1995;91:1659–1668.

21 Maggioni AP, Zuanetti G, Mantini L, et al The predictive value of pre-discharge heart rate

on 8-month mortality in 7,831 patients with acute myocardial infarction in the fibrinolytic era Eur Heart J 1997;18(Abstract suppl):352A.

22 Gordon D, Guyton J, Karnovsky N Intimal alterations in rat aorta induced by stressful stimuli Lab Invest 1983;45:14–19.

23 Mangoni AA, Mircoli L, Giannattasio C, et al Heart rate-dependence of arterial distensibility

in vivo J Hypertens 1996;14:897–901.

24 Palatini P Exercise haemodynamics in the normotensive and the hypertensive subject Clin Sci 1994;87:275–287.

25 Beere PA, Glagov S, Zarins CK Retarding effect of lowered heart rate on coronary sclerosis Science 1984;226:180–182.

athero-26 Bassiouny HS, Zarins CK, Kadowaki MH, et al Hemodynamic stress and experimental iliac atherosclerosis J Vasc Surg 1994;19:426–434.

aorto-27 Kaplan JR, Manuck SB, Clarkson TB The influence of heart rate on coronary artery sclerosis J Cardiovasc Pharmacol 1987;10(Suppl 2):S100–S102.

athero-28 Stamler J, Berkson DM, Dyer A, et al Relationship of multiple variables to blood pressure— findings from four Chicago epidemiologic studies In: Paul O, ed Epidemiology and Control

of Hypertension Symposia Specialists, Miami, 1975, pp 307–352.

29 Cirillo M, Laurenzi M, Trevisan M, et al Hematocrit, blood pressure, and hypertension The Gubbio Population Study Hypertension 1992;20:319–326.

30 Stern MP, Morales PA, Haffner SM, et al Hyperdynamic circulation and the insulin resistance syndrome (“Syndrome X”) Hypertension 1992;20(6):802–808.

31 Palatini P, Julius S Association of tachycardia with morbidity and mortality: ological considerations J Hum Hypertens 1997;11(Suppl 1):19–27.

pathophysi-32 Palatini P, Casiglia E, Pauletto P, et al Relationship of tachycardia with high blood pressure and metabolic abnormalities A study with mixture analysis in three populations Hyperten- sion 1997;30:1267–1273.

33 Julius S, Gudbrandsson T, Jamerson K, et al Hypothesis The hemodynamic link between insulin resistance and hypertension J Hypertens 1991;9:983–986.

34 Julius S, Pascual AV, London R Role of parasympathetic inhibition in the hyperkinetic type

of borderline hypertension Circulation 1971;44:413–418.

35 Deibert DC, DeFronzo RA Epinephrine-induced insulin resistance in man J Clin Invest 1980; 65:717–721.

36 Zeman RJ, Ludemann R, Easton TG, et al Slow to fast alterations in skeletal muscle fibers caused by clenbuterol, a beta-2-receptor agonist Am J Physiol 1988;254:E726–E732.

37 Jamerson KA, Julius S, Gudbrandsson T, et al Reflex sympathetic activation induces acute insulin resistance in the human forearm Hypertension 1993;21(5):618–623.

38 Pollare T, Lithell H, Selinus I, et al Application of prazosin is associated with an increase of insulin sensitivity in obese patients with hypertension Diabetologia 1988;31:415–420.

39 Palatini P, Visentin PA, Mormino P, et al Left ventricular performance in the early stages of systemic hypertension Am J Cardiol 1998;81:418–423.

40 Julius S Altered cardiac responsiveness and regulation in the normal cardiac output type of borderline hypertension Circ Res 1975;36–37(Suppl I):I-199–I-207.

41 Julius S, Li Y, Brant D, et al Neurogenic pressor episodes fail to cause hypertension, but do induce cardiac hypertrophy Hypertension 1989;13:422–429.

42 Palatini P, Maraglino G, Accurso V, et al Impaired left ventricular filling in hypertensive left ventricular hypertrophy as a marker of the presence of an arrhytmogenic substrate Br Heart

J 1995;73:258–262.

43 Palatini P Heart rate as a cardiovascular risk factor Eur Heart J 1999;20(Suppl B):B3–B9.

44 Palatini P Need for a revision of the normal limits of resting heart rate Hypertension 1999; 33:622–625.

45 Coburn AF, Grey RM, Rivera SM Observations on the relation of heart rate, life span, weight and mineralization in the digoxin-treated A/J mouse Johns Hopkins Med J 1971;128:169–193.

46 Kaplan JR, Manuck SB, Adams MR, et al Inhibition of coronary atherosclerosis by propranolol

in behaviorally predisposed monkeys fed an atherogenic diet Circulation 1987;76:1364–1372.

47 Kjekshus JK Importance of heart rate in determining beta-blocker efficacy in acute and term acute myocardial infarction intervention trials Am J Cardiol 1986;57:43F–49F.

long-48 Teo KK, Yusuf S, Furberg CD Effects of prophylactic antiarrhythmic drug therapy in acute myocardial infarction: an overview of results from randomized controlled trials JAMA 1993; 270:1589–1595.

49 The Goteborg Metoprolol Trial in Acute Myocardial Infarction Am J Cardiol 1984;53: 10D–50D.

50 The International Collaborative Study Group Reduction of infarct size with the early use of timolol in acute myocardial infarction N Engl J Med 1984;310:9–15.

51 Taylor SH, Silke B, Ebbutt A, et al A long-term prevention study with oxprenolol in coronary heart disease N Engl J Med 1982;307:1293–1301.

52 Kjekshus JK Comments on beta-blockers: heart rate reduction, a mechanism of action Eur Heart J 1985;6(Suppl A):29–30.

53 MRC Working Party Medical Research Council trial of treatment of hypertension in older adults: principal results Br Med J 1992;304:405–412.

54 Lehtonen A Effect of beta blockers on blood lipid profile Am Heart J 1985;109:1192–1198.

55 Eichorn EJ, Bristow MR Medical therapy can improve the biological properties of the cally failing heart Circulation 1996;94:2285–2296.

chroni-56 Packer M, Bristow MR, Cohn JN, for the U.S Carvedilol Heart Failure Study Group The effect of Carvedilol on morbidity and mortality in patients with chronic heart failure N Engl

phe-59 The Danish Study Group on Verapamil in Myocardial Infarction Effect of verapamil on tality and major events after acute myocardial infarction The Danish Verapamil Infarction Trial II (DAVIT-II) Am J Cardiol 1990;66:779–785.

mor-60 Alderman MH, Cohen H, Rogué R, et al Effect of long-acting and short-acting calcium antagonists on cardiovascular outcomes in hypertensive patients Lancet 1997;349:594–598.

61 The Multicenter Diltiazem Postinfarction Trial Research Group The effect of diltiazem on mortality and reinfarction after myocardial infarction N Engl J Med 1988;319:385–392.

62 Luscher TF, Clozel JP, Noll G Pharmacology of the calcium antagonist mibefradil J tens 1997;15(Suppl 3):S11–S18.

Hyper-63 Kung CF, Tschudi MR, Noll G, et al Differential effects of the calcium antagonist mibefradil

in epicardial and intramyocardial coronary arteries J Cardiovasc Pharmacol 1995;26:312–318.

64 Clozel J, Ertel EA, Ertel SI Discovery and main pharmacological properties of mibefradil (Ro 40-5967), the first selective T-type calcium channel blocker J Hypertens 1997;15(Suppl 5): S17–S25.

65 Van Zwieten PA Centrally acting antihypertensives: a renaissance of interest Mechanisms and haemodynamics J Hypertens 1997;15(Suppl 1):S3–S8.

66 Ernsberger P, Koletsky RJ, Collins LA, et al Sympathetic nervous system in salt-sensitive and obese hypertension: amelioration of multiple abnormalities by a central sympatholitic agent Cardiovasc Drugs Ther 1996;10:275–282.

67 Rosen P, Ohly P, Gleichmann H Experimental benefit of moxonidine on glucose metabolism and insulin secretion in the fructose-fed rat J Hypertens 1997;15(Suppl 1):S31–S38.

68 Ernsberger P, Friedman JE, Koletsky RJ The I1-Imidazoline receptor: from binding site to therapeutic target in cardiovascular disease J Hypertens 1997;15(Suppl 1):S9–S23.

From: Contemporary Cardiology:

Blood Pressure Monitoring in Cardiovascular Medicine and Therapeutics

Edited by: W B White © Humana Press Inc., Totowa, NJ

AMBULATORY BLOOD PRESSURE MONITORING AS A TOOL

ELECTROLYTES AND CIRCADIAN RHYTHMS

NEURO-HUMORAL PATTERNS AND CIRCADIAN BLOOD PRESSURE

cir-BP differs, with reference to external clocktime, if compared to daytime active

individuals The biologic time structure of man is an inherited characteristic for

a number of parameters, such as BP Its normal expression, however, may beinfluenced by either environmental/nutritional factors or an individual’s nor-mal or pathophysiologically acquired neuro-humoral status When normal phaserelationships change between circadian bioperiodicities, BP patterns may alterradically and unpredictably The purpose of this review is to characterize the neuro-humoral and nutritional determinants of the ambulatory blood pressure (ABP)profile in normotensive and hypertensive patients In particular, this review focuses

on the sympathetic nervous system (SNS), the renin–angiotensin–aldosterone(RAA) axis, and the role of dietary sodium (Na+) and potassium (K+) in shapingcircadian BP patterns

AMBULATORY BLOOD PRESSURE MONITORING AS A TOOL

Ambulatory BP monitoring is a recently developed methodology, capable ofidentifying and systematically evaluating factors responsible for individual dif-ferences in BP responses in the natural environment This approach provides ameans for studying an individual in a standardized fashion as he or she responds

to the physical and psychological demands of a typical 24-h day Prior researchemploying ABP monitoring indicates that most people display low-amplitudediurnal variations in BP, with higher pressures during waking hours and lower

pressures during sleep (1,2) In most normotensive subjects, average BP values decline by approx 15% during sleep (3–5) In hypertensive subjects, the circa-

dian rhythm is generally preserved, although the 24-h BP profile shifts to higher

around-the-clock values (6).

Ambulatory BP patterns are rarely static with considerable day-to-day

vari-ability in how nocturnal BP patterns express themselves (7) It has proven

tempt-ing to assign causality to a particular dietary or neuro-humoral change in hownocturnal BP changes occur Unfortunately, it is the rare circumstance where aspecific neuro-humoral or dietary pattern is exclusively responsible for a par-ticular nocturnal BP pattern, such as nondipping (minimal drop in nocturnal BP).Rather, factors typically coalesce with different weightings assigned to indi-vidual factors in order to arrive at a final explanation for a specific BP pattern.Comments found in this chapter should be viewed accordingly

ELECTROLYTES AND CIRCADIAN RHYTHMS

The established associations between BP and electrolytes are for the most partmost reliable when based on data from urinary excretion and/or a validated self-

report of nutrient intake (7–9) Urinary excretion and/or dietary recall parameters

are the preferred correlates to BP, as it is widely held that they more realistically

depict the true state of electrolyte balance Interpreting the relationship between

a plasma electrolyte, such as K+, and BP is inherently difficult because tional intake is one of only many factors known to influence plasma K+values.Such factors include a circadian rhythm for plasma K+(average peak z troughdifferenceF 0.60 mEq/L with lowest values at night) (10) and a tendency for K+

nutri-to migrate intracellularly, when `2-adrenergic receptors are stimulated (11).

Accordingly, very few reports have even attempted to characterize the tionship between plasma K+and ABP patterns in hypertensive patients (12,13).

rela-Goto et al found significant negative correlations between daytime plasma K+concentration and 24-h systolic and diastolic BP levels in patients with essential

hypertension (13) Plasma K+also inversely correlated with both daytime andnighttime systolic and diastolic BP levels In these studies, there was no corre-lation between office BP readings and plasma K+concentration No doubt, anysuch relationship was obscured by the inherent variability of office BP measure-ments (Fig 1)

If the plasma K+value in any way equates with intake, these results are sistent with prior epidemiologic studies, which have found a negative correlationbetween K+intake and BP levels (12) Goto et al have further suggested that

con-decreased extracellular K+promotes vasoconstriction in hypertensive patients

by either enhancing SNS activity or by increasing the Na+content of vascular

smooth muscle cells (13) Additional research is needed to better understand the

relative contribution of plasma electrolytes to circadian variability in BP

Fig 1 Relation between plasma K+and 24-h systolic blood pressure (A: r = 0.336, p < 0.01)

or office systolic blood pressure (B: r = <0.018, p = NS) in 82 patients with essential tension Adapted with permission of Elsevier Science from ref (13) Copyright 1997 by

hyper-American Journal of Hypertension Ltd.

NEURO-HUMORAL PATTERNS AND CIRCADIAN BLOOD PRESSURE RHYTHMS

Atrial Natriuretic Peptide

As a prominent regulatory arm of volume homeostasis in man, the natriureticpeptides are intimately involved in the regulation of BP Atrial natriuretic pep-tide (ANP) release is primarily regulated by atrial pressure though a number ofother factors, such as age and level of renal and/or cardiac function, that canarbitrate the final plasma concentration for ANP ANP can be viewed as the

“mirror image” of the RAA axis in that it inhibits the release of renin and rone while opposing the actions of angiotensin II and aldosterone through effects

aldoste-on vascular taldoste-one, cells growth, and renal sodium reabsorptialdoste-on When ANP isadministered to animals or humans, the BP acutely drops, a process, which isparticularly prominent when the RAA is activated For these reasons, a relation-ship between the time structure of ANP, other neurohormones, and 24-h BPpatterns has been sought

It has been observed that single-point-in-time morning ANP levels may have

either no relationship to 24-h BP (14) or may separate isolated clinic

hyperten-sion (wherein ANP levels are typically normal) from sustained hypertenhyperten-sion

(wherein plasma ANP levels are increased) in elderly hypertensives (15)

Meth-odologic considerations are important to the interpretation of circadian ANPpatterns For example, Chiang et al observed the absence of any circadian rhythmfor ANP (and thus no relationship to diurnal BP change) in a group of 14 healthy

volunteers in whom ANP was sampled every 3 h for 24 h (16).

In other studies in which subjects were synchronized to the light–dark cycle andwere given a controlled diet, a variable acrophase for ANP was found Portaluppi

et al originally noted an acrophase for ANP to occur at around 4:00 AM In thesestudies, BP and heart rate (HR) rhythms appeared to be in antiphase with the ANPrhythm, with the peak of BP and HR more or less coinciding with the troughfor ANP rhythm This pattern of response suggested a relationship between ANP

levels and BP and HR (17) Alternatively, Cugini et al noted an acrophase timing

for ANP at about 5:00 PMin young clinically healthy subjects and no circadianpattern for ANP in elderly subjects, although mean blood levels of ANP were

noticeably higher in the elderly cohort (18) There is no obvious explanation for

these obviously different findings Additional studies will be required to clarifythe time pattern of ANP levels

Plasma Renin Activity

Gordon et al originally described a diurnal rhythm for plasma renin activity

(PRA) that was independent of posture and dietary influences (19), a finding subsequently corroborated by a number of other investigators (17,20–22) From

these observations emerged the concept of a circadian rhythm in PRA, with a

nadir in the afternoon and a nocturnal increase culminating in the early morninghours, despite the occasional study having failed to demonstrate any significant

variation in PRA with the time of day (23–25) What remained to be determined

was to define the relative role of endogenous circadian rhythmicity and the sleep–wake cycle on 24-h PRA variations because sleep can make substantial contribu-tions to the overall variations in PRA and thereby mask the characteristics of an

endogenous rhythm (26).

A strong relationship exists between nocturnal oscillations in PRA and

inter-nal sleep structure (24,27) Non-rapid-eye-movement (NREM) is invariably

linked to increasing PRA levels, with PRA decreasing during ment (REM) sleep In normal man, modifying the renal renin content modulatesonly the amplitude of the nocturnal oscillations without altering their relation-

rapid-eye-move-ship to the stage of sleep (28), and in the case of sleep disorders, such as sleep apnea, the PRA profiles reflect all facets of the sleep structure disturbance (29).

Brandenberger et al have recently shown, using an acute shift in the normal sleeptime, that increased renin release was associated with sleep whatever time it

occurs, an observation atypical of an intrinsic circadian rhythm (Fig 2) (26).

This group further observed that internal sleep architecture had an importantmodulatory role on the characteristics of the PRA oscillations and, consequently,

on the 24-h pattern When NREM–REM sleep cycles are disturbed, as is the casewith the fragmented sleep of obstructive sleep apnea, there is insufficient timefor PRA to increase significantly; consequently, in poor sleepers, PRA valuesmay not vary to any significant degree throughout a 24-h time span This may pro-vide an explanation for the occasional study wherein PRA values fail to increase

during sleep (23–25,30).

Although several studies have examined the 24-h cycle of PRA, few have seenfit to examine the relationship between PRA and ABP patterns, and those thathave, fail to provide a consistent picture For example, Watson found significantpositive correlations between PRA and variability in daytime BP readings after

adjusting for age (31) Chau et al (32), however, reported negative correlations between upright PRA and 24-h mean BP readings Harshfield and colleagues (33)

examined the relationship between renin–sodium profiles and ABP patterns inhealthy children The subjects were classified as low, intermediate, or high renin,inferred from the relationship between PRA and Na+excretion The subjectsclassified as high renin had elevated systolic and diastolic BP readings whileasleep more so than did subjects in the low-renin category These studies suggestthat the relationship between the level of RAA system activity and ambulatory

BP patterns is complex, with Na+sensitivity and/or Na+intake emerging as tant co-variables in this relationship

impor-In addition, the 24-h pattern for PRA oppose that for BP, which tends to fall

in the first few hours of sleep and to rise thereafter Superimposed on these dencies, periodic changes in BP occur that coincide with the presence of NREM

ten-sleep cycles (34) Such changes are characterized by slight decreases in the mean

BP levels during slow-wave sleep and small increases in mean BP levels duringREM sleep, during which there is a marked increase in PRA pulse activity It isunclear as to the relationship of PRA pulse activity to these observed oscillations

in nocturnal BP

Angiotensin-Converting Enzyme Inhibitors

A failure to establish clear relationships between nocturnal RAA axis activityand BP patterns may be a function of various sensitivities to external influences

Fig 2 Effects of an 8-h delay of the sleep–wake cycle on the 24-h plasma renin activity

profiles in 10 subjects: (A) normal nocturnal sleep from 2300 to 0700 h and (B) daytime

sleep from 0700 to 1500 h after a night of sleep deprivation Values are expressed as means

± SEM Adapted from (26) by permission of Lippincott Williams & Wilkins.

and/or the interactions of other rhythms, which could obscure PRA cycles duringthe waking periods One way to evaluate the importance of nocturnal PRA is todetermine the nature of the vasodepressor response subsequent to administration

of ACE inhibitors Several studies have attempted such an evaluation (Table 1).Possible circadian changes in the pharmacokinetics and effect on serumangiotensin-converting enzyme (ACE) activity of the ACE inhibitor enalapril

were first evaluated in the studies of Weisser et al (35), with several subsequent studies having been reported since that time (Table 1) Weisser et al (35) noted

that the mean serum concentration–time profiles of enalapril and its activemetabolite enalaprilat were comparable whether enalapril was ingested at 0800

or 2000 h Administration of enalapril at 2000 h did not markedly influence thebioavailability of enalapril as estimated by time to maximum concentration

(Tmax), maximum drug concentration (Cmax), or area under the curve (AUC0-24)for the active enalapril metabolite, enalaprilat The only observed difference

was an increase in Tmaxfor enalapril after its evening administration (1.3 ± 0.5[0800 h]) vs 2.4 ± 1.4 (2000 h [p < 0.05]), a phenomenon that has been observedwith a number of other drugs

Palatini et al subsequently examined the relationship between daytime (0800 h)and nighttime (2200 h) administration of quinapril following 4 wk of dosing

(36) The 24-h BP profiles obtained by ABP monitoring showed a more sustained

Table 1 Chronopharmacology of ACE inhibitors

Single/

Authors Drug/dose Subject No Timing multiple

Weisser Enalapril/10 8 0800 Yes/no Tmaxw with 2000 h dose

Palatini Quinapril/20 18 0800 No/yes Evening dose maintained

double-blind/ dose lost efficacy during

Palatini Benazepril/10 10 0900 Yes/no Morning had more sustained

crossover

double blind/ morning dose lost efficacy

Morgan Perindopril/4 20 0900 No/yes Evening dose maintained

crossover dose lost efficacy after 18 h

antihypertensive action with the evening administration (2200 h) of quinaprilcompared with its morning administration (0800 h) There was a partial loss ofeffectiveness for quinapril during the nighttime and early morning hours when

it was administered in the morning In addition, measurement of ACE activityshowed that evening administration of quinapril caused a less pronounced butmore sustained decline of plasma ACE In this regard, 24 h after the last dose

of quinapril, the residual ACE inhibition was greater (62%) with evening ing than was the case with morning dosing (40%) These authors concluded thatevening administration of quinapril was preferable because it provided a morehomogeneous pattern of 24-h BP control, which, in part, may have related to an

dos-extended inhibition of the ACE enzyme (36).

Palatini et al also evaluated the influence of timing of benazepril tion on 24-h intra-arterial BP measurements In contradistinction to their previ-ous studies with quinapril, they noted that a single 10-mg dose of benazepriladministered at 0900 h more effectively covered the 24-h dosing interval than did

administra-an identical dose administered at 2100 h (37) Although the single-dose nature

of these studies make their interpretation difficult, they do suggest that ACEinhibitor pharmacokinetics are relevant to the circadian variability in response

to an ACE inhibitor

Witte et al (38) evaluated the cardiovascular effects and pharmacokinetics of

once-daily enalapril (10 mg) after either a single dose or following its chronicadministration Chronic therapy (with dosing at 0700 h) significantly reduced

BP during the day but lost effectiveness between 2 and 7 AMof the succeedingday Chronic dosing at 1900 h significantly exaggerated the nocturnal dip in BP

BP values slowly increased throughout the next day with the evening dosingregimen, with no effect on elevated afternoon values Peak concentrations ofenalaprilat were found at 3.5 h (morning) and 5.6 h (evening) after drug admin-istration The time-to-peak drug effect was shorter after morning dosing (7.4 ±4.3 h [diastolic]) than evening dosing (12.1 ± 3.7 h [diastolic]) Differences inthe response to enalapril could not be attributed to timewise changes in pharma-cokinetics or to a different time-course of ACE inhibition It is more likely thatcircadian changes in the sensitivity of the RAA system play an important role indefining timewise differences in response to an ACE inhibitor

In a final study, Morgan et al examined the BP response following the istration of perindopril either in the morning (0900 h) or in the evening (2100 h)

admin-It was noted in these studies that the early morning rise in BP was reduced morewith the evening administration of perindopril However, the 2100 h dose regi-men did not reduce BP over 24 h, whereas the 0900 h dose achieved better BPcontrol These studies concluded that the time-related response profile obtainedwith an ACE inhibitor is unique and that chronobiology has important effects on

the action of these drugs (39).

Plasma Aldosterone

Plasma aldosterone secretion follows a pattern such that mean hormone

con-centrations are highest during the night and early morning (20–23,40) Plasma

aldosterone values during a 24-h time period appear to be coupled to PRA, withrenin secretion being either simultaneous with or preceding aldosterone secre-tion by 10–20 min, with this temporal coupling enhanced in a low-sodium state

(40) Under basal conditions, the relative contribution of sleep processes and

circadian rhythm to plasma aldosterone levels remains poorly defined, larly as relates to those systems that cojointly control aldosterone release (renin-angiotensin, adrenocorticotropic, and dopaminergic systems)

particu-Heretofore, any timewise change in the 24-h profile of aldosterone was viewedsimplistically as a circadian event More recently, it has been recognized that the

pattern of aldosterone release is influenced by sleep architecture (41) Recent

stud-ies, employing an experimental design of abruptly shifting sleep by 8 h, show sleepprocesses to have a stimulatory effect on aldosterone release, as demonstrated

by high mean levels together with high pulse amplitude and pulse frequencyobserved during the sleep period and reduced levels during sleep deprivation

(Fig 3) This pattern of secretion is similar to that observed with PRA (26) The

large increase in plasma aldosterone levels and pulse amplitude following ening from nocturnal sleep is attributable to an increase in activity of the adreno-corticotropic axis, reflected by the surge in cortisol in the early morning Theissue of nocturnal aldosterone change is complex, with aldosterone pulses mainly

awak-Fig 3 Effect of an 8-h shift in sleep period cycle on 24-h profiles for plasma aldosterone

in seven subjects Blood was sampled at 10-min intervals In the daytime sleep condition, the amplitude of the aldosterone pulses was significantly enhanced during the sleep period Values are expressed as mean ± SEM Adapted with permission (41).

related to PRA oscillations during the sleep periods, whereas aldosterone pulsesare associated with cortisol pulses during the waking periods.

The influence of aldosterone circadian patterns on BP and, in particular,nocturnal BP is poorly defined Little meaningful information exists that mightpermit an assessment of the role of aldosterone antagonism in modifying circa-dian BP patterns

Sympathetic Nervous System

In both normotensive and hypertensive individuals, the BP fluctuates ing to the level of both mental and physical activities BP, HR, and SNS activ-ity are typically highest when a hypertensive patient is awake and/or active.Conversely, these values reach a nadir between midnight and 3:00 AM(42–44).

accord-Although the exact interplay of all physiologic and pathophysiologic mediators

of the diurnal rhythm remains unclear, nocturnal BP and HR seems to track SNSactivity best—but not entirely so Experiments with autonomic blocking agentsprovide some insight into the importance of the SNS in diurnal BP rhythms Forexample, the BP rhythm in high spinal cord transected patients (with completetetraplegia) is nonexistent, despite HR variability being preserved (presumably

because cardiac vagal innervation remains intact) (45) Paraplegics and

incom-plete tetraplegics typically have a normal diurnal BP pattern These findings areconsistent with the thesis that central SNS outflow is an important determinant

of the normal diurnal rhythm of BP

Attempts to define the role of the SNS in determining nocturnal BP changesare complicated by methodologic constraints This being said, SNS activity

typically diminishes while asleep, with changes in the sympathoadrenal branch

(epinephrine) being governed in a dual fashion by both posture and sleep and the

noradrenergic branch (norepinephrine) being regulated more so by posture (44).

Diurnal changes in plasma catecholamine values, as markers of SNS activity, aresubject to considerable sampling error and require careful interpretation as to thestudy conditions under which they were obtained Plasma epinephrine concen-trations and/or SNS activity decline during sleep (particularly during NREM

sleep) and begin to increase in conjunction with morning awakening (44,46,47) and/or episodically during episodes of REM sleep (Fig 4) (48) Plasma norepi-

nephrine concentrations trend downward when asleep and do not significantlyincrease until a postural stimulus to norepinephrine release, such as the upright

position, is added to changes accompanying the arousal process (44,46)

Morn-ing plasma norepinephrine concentrations, although typically higher than sleepvalues, are not necessarily the highest values attained during a 24-h time interval

(46,47) Finally, microneurography, a specific marker of muscle SNS activity,

fails to show any increased neural activity in normal volunteers when performedbetween the hours of 6:30 AMand 8:30 AM, a time parenthetically when the rate

of myocardial infarction is highest This suggests that the early morning peak in

myocardial infarction and/or sudden cardiac death could, in part, reflect gerated end-organ responsiveness to norepinephrine following the relative sym-

exag-pathetic withdrawal that occurs during sleep (49).

Nocturnal BP can assume a number of different and now well-characterizedpatterns: extreme dipping (an approximate 30% ¦ in BP while asleep), normaldipping (a 10–20% ¦ in nighttime BP), and nondipping (minimal drop in noctur-

nal BP or a rise in BP at night) (7,50) Of these BP patterns, attention has recently

centered on the significance of a nighttime nondipping BP pattern, because it is

Fig 4 Recordings of sympathetic nerve activity (SNA) and mean blood pressure (BP) in

a single subject while awake and while in stages 2, 3, and 4 and REM sleep As non-REM sleep deepens (stages 2–4), SNA gradually decreases and BP (mmHg) and variability in

BP are gradually reduced Arousal stimuli elicited K complexes on the electrocardiogram (not shown) were accompanied by increases in SNA and BP (indicated by the arrows, stage

2 sleep) In contrast to the changes during non-REM sleep, heart rate, BP, and BP ity increased during REM sleep, together with a profound increase in both the frequency and amplitude of SNA There was a frequent association between REM twitches (momen- tary periods of restoration of muscle tone, denoted by T on the tracing) and abrupt inhibi-

variabil-tion of SNA and increases in BP Adapted with permission (48) Copyright 1993

Massachu-setts Medical Society.

believed to be associated with more rapid progression of renal failure (51) and/

or a greater degree of left ventricular hypertrophy (52) Aging, salt sensitivity,

and African-American ethnicity are viewed as relevant demographic markers

for this phenomenon (7).

Little is known about the pathophysiology of nocturnal nondipping in eithernormotensive or essential hypertensives, although important clues to the origin

of this phenomenon can be extracted from an analysis of sleep patterns Sleeparchitecture and SNS activity are important determinants of nocturnal BP and

HR During NREM sleep, there is a tendency for HR to slow and BP to fall, aprocess characterized by a relative increase in parasympathetic or vagal activ-

ity (53–55) It is now fairly well accepted that alterations in SNS activity may

lead to relevant effects on the pathophysiology of sleep, as well as influence thediurnal BP profile Derangements in autonomic nervous system activity, sleep-disordered breathing, and alterations in sleep architecture and duration are well-

recognized causes of change in the circadian BP profile (54) In addition, sleep

disturbances are reported to influence the circadian BP profile Schillaci et al.showed that the reported duration of sleep was significantly shorter for hyper-

tensive “nondippers” than it was for “dippers” both in males and females (56).

Kario et al found nondippers to have increased nocturnal physical activity, as

determined by actigraphy (57) Thus, the duration and quality of sleep should be

considered in the interpretation of the diurnal BP profile

Nutrition

The intake of Na+and/or K+is an important modulator of BP The impact ofsuch nutritional modification has most typically been assessed by evaluating

change in casual BP determinations (58,59), although more recently, ABP

tech-nology has been employed to delineate the 24-h pattern of change with such

inter-ventions (60–64) Accordingly, it is only in the last decade that nocturnal BP patterns could serve as targets for dietary intervention (60,64).

Prior research has identified demographic groups in whom the equilibriumpoint for Na+balance is set at a higher level of BP For example, Weinberger et

al demonstrated that blacks and older individuals (>40 yr) poorly excrete a Na+load, and in order to achieve Na+balance, higher BP values are required for a

longer period of time (65) Falkner et al have also reported that salt-sensitive

adolescents with a positive family history of hypertension had greater increases

in BP with salt loading than did adolescents who were either salt resistant or had

a negative family history of hypertension (66) Harshfield et al have also

demon-strated that Na+ intake is an important determinant of ABP profiles in black

children and adolescents (67) Black subjects displayed a positive correlation

between Na+excretion and asleep systolic BP, whereas Na+excretion was pendent of asleep BP in white subjects

inde-Several investigators have probed the relationship between salt sensitivity andthe nocturnal decline in ABP Wilson et al examined the relationship between

salt sensitivity and ABP in healthy black adolescents (62) They classified 30%

of those studied as salt sensitive according to predetermined criteria for saltsensitivity, with the remaining subjects designated as salt resistant Salt-sensi-tive subjects showed higher daytime diastolic and mean BP than did salt-resis-tant subjects A significantly greater percentage of salt-sensitive subjects wereclassified as nondippers according to diastolic BP (<10% decrease in BP fromawake to asleep) as compared to salt-resistant individuals (Fig 5) These resultswere some of the first to indicate that salt sensitivity is associated with a nondippernocturnal BP pattern in healthy black adolescents These findings are consistent

with prior observations by de la Sierra et al (63), which showed higher awake

BP values in normotensive salt-sensitive adults as compared to salt-resistant adults,and a recent meta-analysis that found American blacks to experience a smaller

dip in BP (higher levels of both systolic and diastolic BP) at night (68).

The mechanism(s) by which Na+sensitivity (or sodium loading) alters nal BP (although incompletely elucidated) likely involves increased SNS activ-

noctur-ity (65,69,70) Increased SNS activnoctur-ity, in turn, is known to modify Na+handling,albeit in a mixed fashion For example, Harshfield et al have found that normo-tensive individuals differ in Na+handling during SNS arousal (71) In one group

Fig 5 Percentage of salt-sensitive versus salt-resistant normotensive adolescent blacks

who were classified as dippers (>10% decline in nocturnal blood pressure) or nondippers (<10% decline in nocturnal blood pressure) Adapted with permission of Elsevier Science

from ref (62) Copyright 1999 by American Journal of Hypertension Ltd.

of adults, termed excreters, Na+excretion increased during 1 h of behaviorallyinduced SNS arousal (competitive video games) with a return to baseline levels

within 2 h of stimulation In a second group of adults, termed retainers, Na+tion decreased in response to SNS arousal and remained below baseline values

excre-for at least 2 h following stimulation The findings of Harshfield et al (71), as

well as those of several other investigators, now suggest an important interactiverole for SNS activation in Na+retention and, by this process, a means by which

nocturnal BP might be altered (72,73).

The role of Na+intake in the definition of nocturnal BP is evident from the

studies of Uzu et al (60) and Higashi et al (61) Uzu et al found that a nondipper

nocturnal BP pattern in salt-sensitive patients converted to a dipper pattern with

Na+restriction (Fig 6) (60) Higashi et al (61) found that the nocturnal decline

in mean BP was significantly smaller in salt-sensitive as compared to tant hypertensives during a Na+-loading protocol, sufficient to have elevatedABP levels (Fig 7) In their studies, nondipping was also most commonly seen

salt-resis-in those hypertensive patients who were salt sensitive and exposed to a high-Na+diet These findings suggest that a high-Na+intake can now be considered as an

Fig 6 Relationships of changes in nocturnal mean arterial pressure (MAP) fall induced

by Na + restriction with the sodium-sensitivity index as well as with nocturnal MAP fall before Na + restriction The sodium-sensitivity index, shown on the left, was calculated as the ratio of the change in MAP over the change in urine sodium excretion (UNaV) produced

by Na + restriction (1–3 g NaCl/d) The nocturnal fall in MAP before Na + restriction (on the right) was calculated as the difference between daytime and nighttime MAPs during high

Na + intake (12–15 g NaCl/d) The change in nocturnal MAP fall with Na + restriction was calculated as the difference between low- and high-Na + diets and had a positive relation-

ship with the sodium-sensitivity index (r = 0.38, p < 0.02) and a negative relationship with

the nocturnal MAP fall during the high-Na +diet (r = <0.75, p < 0.0001) Adapted with permission (60).

etiologic factor (among several others) for the failure of BP to decline at night

in hypertensive patients, particularly in those who are salt sensitive

Fewer investigations have examined the relationship between K+intake and

ABP responses (7,64) A potential pathway by which K+may influence ABPpatterns involves K+-related natriuresis For example, a number of studies havestrongly suggested that a change in K+intake alters Na+balance, such that dietary

K+restriction results in Na+retention and K+supplementation produces a

natri-uretic response (74,75) The effect of K+on urinary Na+excretion, plasma volume,and mean arterial pressure could also be evidence for a K+-mediated vasodilatoreffect on BP For example, it is well established that the local intra-arterial infu-sion of K+decreases forearm vascular resistance and increases forearm blood

flow in a dose-dependent fashion (76,77) It has also been demonstrated that

K+supplementation given in combination with a high-Na+diet suppresses theincrease in catecholamines, which typically occurs in response to Na+loading (70).

Theoretical premises such as these led Wilson et al to examine the effects of

a 3-wk increase in K+on ABP responses in healthy black adolescents Subjectswere classified as dippers or nondippers according to whether they sustained

a >10% decrease from awake to asleep BP, and were randomized to either ahigh-K+diet or a usual diet control group A significant proportion of nondippersswitched from a nondipper to dipper status in response to the high-potassium diet

Fig 7 Scatterplot showing the relationship between the nocturnal decline in blood

pres-sure during a high-NaCl diet (340 mmol/d) (nocturnal BP decline) and the NaCl-induced increase in blood pressure (salt sensitivity) The NaCl-induced increase in BP was corre- lated with the nocturnal decline in BP during a high-NaCl diet but not during a low-NaCl

diet Adapted with permission (61).

(7) Although this study did not show a change in nocturnal BP, a subsequent

study, which specifically examined for this effect in salt-sensitive subjects, didshow a reversal in nighttime BP as a consequence of a high-K+diet in salt-sens-

itive individuals (64) These studies strongly suggest that K+intake may have asubstantial influence on diurnal patterns of BP responses

SUMMARY

How a nocturnal BP pattern presents is a consequence of both intrinsic dian rhythms and the quantity and quality of sleep Although a range of neuro-humoral factors can influence the circadian BP pattern, abnormal SNS activity

circa-is most commonly linked to a dcirca-isappearance of the normal decline in nocturnal

BP Nutritional intake, such as either a high-Na+or a low-K+intake, can alsoerase the normal decline in nocturnal BP The impact of nutrition on nocturnal

BP change is most prominent in salt-sensitive individuals Additional studies of

an integrative nature will be necessary to more completely define the dynamicinterplay between nutrition and various neuro-humoral axes in how nocturnal

BP patterns are expressed

3 Pickering TG Sleep, circadian rhythms and cardiovascular disease Cardiovasc Rev Rep 1980;1:37–47.

4 Staessen JA, Fagard RH, Lijnen PJ, et al Mean and range of ambulatory blood pressure in normotensive subjects from a meta-analysis of 23 studies Am J Cardiol 1991;67:723–727.

5 Staessen JA, Bieniaszewski L, O’Brien E, et al Nocturnal blood pressure fall on ambulatory monitoring in a large international base Hypertension 1997;27:30–39.

6 Fogari R, Ambrosoli S, Corradi L, et al 24-Hour blood pressure control by once-daily istration of irbesartan assessed by ambulatory blood pressure monitoring J Hypertens 1997;15: 1511–1518.

admin-7 Wilson DK, Sica DA, Devens M, et al The influence of potassium intake on dipper and nondipper blood pressure status in an African-American adolescent population Blood Pres- sure Monitor 1996;1:447–455.

8 Gill JR, Gullner HG, Lake CR, et al Plasma and urinary catecholamines in salt-sensitive idiopathic hypertension Hypertension 1988;11:312–319.

9 Staessen JA, Birkenhager W, Bulpitt CJ, et al The relationship between blood pressure and sodium and potassium excretion during the day and night J Hypertens 1003;11:443–447.

10 Solomon R, Weinberg MS, Dubey A The diurnal rhythm of plasma potassium: relationship

to diuretic therapy J Cardiovasc Pharmacol 1991;17:854–859.

11 Struthers AD, Reid JL, Whitesmith R, et al Effect of intravenous adrenaline on diogram, blood pressure and plasma potassium Br Heart J 1983;49:90–93.

electrocar-12 Bulpitt CJ, Shipley MJ, Semmence A Blood pressure and plasma sodium and potassium Clin Sci 1981;61:85s–87s.

13 Goto A, Yamada K, Nagoshi H, et al Relation of 24-h ambulatory blood pressure with plasma potassium in essential hypertension J Hypertens 1997;10:337–340.

14 Guasti L, Grimoldi P, Ceriani L, et al Atrial natriuretic peptide and 24 h blood pressure Blood Pressure Monitor 1997;2:89–92.

15 Kario K, Nishikimi T, Yoshihara F, et al Plasma levels of natriuretic peptides and medullin in elderly hypertensive patients: relationship to 24-h blood pressure J Hypertens 1998;16:1253–1259.

adreno-16 Chiang FT, Tseng CD, Hsu KL, et al Circadian variation of atrial natriuretic peptide in normal people and its relationship to arterial blood pressure, plasma renin activity and aldosterone level Int J Cardiol 1994;46:229–233.

17 Portaluppi F, Bagni B, degli Uberti E, et al Circadian rhythms of atrial natriuretic peptide, renin, aldosterone, cortisol, blood pressure and heart rate in normal and hypertensive subjects.

22 Modlinger RS, Sharif-Zadeh K, Ertel NH, et al The circadian rhythm of renin J Clin crinol Metab 1976;43:1276–1282.

Endo-23 Lightman SL, James VHT, Linsell C, et al Studies of diurnal changes in plasma renin activity, and plasma noradrenaline, aldosterone and cortisol concentration in man Clin Endocrinol (Oxf) 1981;14:213–223.

24 Mullen PE, James VHT, Lightman C, et al A relationship between plasma renin activity and the rapid eye movement phase of sleep in man J Clin Endocrinol Metab 1980;50:466–469.

25 Brandenberger G, Follenius M, Nuzet A, et al Ultradian oscillations in plasma renin activity: their relationships to meals and sleep stages J Clin Endocrinol Metab 1985;61:280–284.

26 Brandenberger G, Follenius M, Goichot B, et al Twenty-four-hour profiles of plasma renin activity in relation to the sleep-wake cycle J Hypertens 1994;12:277–283.

27 Brandenberger G, Follenius M, Simon C, et al Nocturnal oscillations in plasma renin activity and REM–NREM sleep cycles in humans: a common regulatory mechanism Sleep 1988; 11:242–250.

28 Brandenberger G, Krauth MO, Erhart J, et al Modulation of episodic renin release during sleep in humans Hypertension 1990;15:370–375.

29 Follenius M, Krieger J, Krauth MO, et al Obstructive sleep apnea treatment: peripheral and central effects of plasma renin activity and aldosterone Sleep 1991;14:211–217.

30 Kool MJ, Wijnen JA, Derkx FH, et al Diurnal variation in prorenin in relation to other humoral factors and hemodynamics Am J Hypertens 1994;7:723–730.

31 Watson RDS, Stallard TJ, Flinn RM, et al Factors determining direct arterial pressure and its variability in hypertensive men Hypertension 1980;2:333–341.

32 Chau NP, Chanudet X, Larroque P Inverse relationship between upright plasma renin ity and 24-hour blood pressure variability in borderline hypertension J Hypertens 1990;8: 913–918.

activ-33 Harshfield GA, Pulliam DA, Alpert BS, et al Renin–sodium profiles influence ambulatory blood pressure patterns in children and adolescents Pediatrics 1990;87:94–100.

34 DeLeeuw PW, Van Leeuwen SJ, Birkenhager WH Effect of sleep on blood pressure and its correlates Clin Exp Hypertens A 1985;7:179–186.

35 Weisser K, Schloos J, Lehmann K, et al Pharmacokinetics and converting enzyme inhibition after morning and evening administration of oral enalapril to healthy subjects Eur J Clin Phar- macol 1991;40:95–99.

36 Palatini P, Racioppa A, Raule G, et al Effect of timing of administration on the plasma ACE inhibitory activity and the antihypertensive effect of quinapril Clin Pharmacol Ther 1992; 52:378–383.

37 Palatini P, Mos L, Motolese M, et al Effect of evening versus morning benazepril on 24-hour blood pressure: a comparative study with continuous intraarterial monitoring Int J Clin Phar- macol Toxicol 1993;31:295–300.

38 Witte K, Weisser K, Neubeck M, et al Cardiovascular effects pharmacokinetics, and ing enzyme inhibition of enalapril after morning versus evening administration Clin Pharma- col Ther 1993;54:177–186.

convert-39 Morgan T, Anderson A, Jones E The effect on 24 h blood pressure control of an angiotensin converting enzyme inhibitor (perindopril) administered in the morning or at night J Hypertens 1997;15:205–211.

40 Vieweg WV, Veldhuis JD, Carey RM Temporal pattern of renin and aldosterone secretion in men: effects of sodium balance Am J Physiol 1992;31:F871–F877.

41 Charloux A, Gronfier C, Lonsdorfer-Wolf E, et al Aldosterone release during the sleep-wake cycle in humans Am J Physiol 1999;276:E43–E49.

42 Pickering TG, James GD Determinants and consequences of the diurnal rhythm of blood pressure Am J Hypertens 1993;6:166S–169S.

43 Millar-Craig MW, Bishop CN, Raftery EB Circadian variation of blood pressure Lancet 1978;1:795–797.

44 Dodt C, Breckling U, Derad I, et al Plasma epinephrine and norepinephrine concentrations

of healthy humans associated with nighttime sleep and morning arousal Hypertension 1997; 30:71–76.

45 Nitsche B, Perschak H, Curt A, et al Loss of circadian blood pressure variability in complete tetraplegia J Hum Hypertens 1996;10:311–317.

46 Linsell CR, Lightman SL, Mullen PE, et al Circadian rhythm of epinephrine and rine in man J Clin Endocrinol Metab 1985;60:1210–1215.

norepineph-47 Stene M, Panagiotis N, Tuck MI, et al Plasma norepinephrine levels are influenced by sodium intake, glucocorticoid administration, and circadian changes in normal man J Clin Endocrinol Metab 1980;51:1340–1345

48 Somers VK, Dyken ME, Mark AL, et al Sympathetic nerve activity during sleep in normal subjects N Engl J Med 1993;328:303–307.

49 Middlekauff HR, Sontz EM Morning sympathetic nerve activity is not increased in humans Implications for mechanisms underlying the circadian pattern of cardiac risk Circulation 1995; 91:2549–2555.

50 Fernandez A, Sica DA, Wilson DK, Gehr TWB, White W Extreme dipping—a definitional dilemma Am J Hypertens 1999;12:150A.

51 Bianchi S, Bigazzi R, Campese VM Altered circadian blood pressure profile and renal damage Blood Pressure Monitor 1997;6:339–346.

52 Palatini P, Penzo M, Racioppa A Clinical relevance of nighttime blood pressure and of time blood pressure variability Arch Intern Med 1992;152:1855–1860.

day-53 Billman GE, Dujardin JP Dynamic changes in cardiac vagal tone as measured by time-series analysis Am J Physiol 1990;258:H896-H902.

54 Portaluppi F, Cortelli P, Provini F Alterations of sleep and circadian blood pressure profile Blood Pressure Monitor 1997;6:301–313.

55 Baumgart P Circadian rhythm of blood pressure: internal and external time triggers biol Int 1991;8:444–450.

Chrono-56 Schillaci G, Verdecchia P, Borgioni C, et al Predictors of diurnal blood pressure changes in

2042 subjects with essential hypertension J Hypertens 1996;14:1167–1173.

57 Kario K, Pickering TG, Schwartz JE Physical activity as a determinant of diurnal blood pressure variation Am J Hypertens 1999;12:31A.

58 Moore TJ, Malarick C, Olmedo A, et al Salt restriction lowers resting blood pressure but not 24-h ambulatory blood pressure Am J Hypertens 1991;4:410–415.

59 Elliott P, Stamler J, Nichols R, et al Intersalt revisited: further analyses of 24-hour sodium excretion and blood pressure within and across populations Br Med J 1996;312:1249–1253.

60 Uzu T, Ishikawa K, Fujita T, et al Sodium restriction shifts circadian rhythm of blood pressure from nondipper to dipper in essential hypertension Circulation 1997;96:1859–1862.

61 Higashi Y, Oshima T, Ozono R, et al Nocturnal decline in blood pressure is attenuated by NaCl loading in salt-sensitive patients with essential hypertension Hypertension 1997;30(Part 1): 163–167.

62 Wilson DK, Sica DA, Miller SB Ambulatory blood pressure non-dipping status in tive and salt-resistant black adolescents Am J Hypertens 1999;12:159–165.

salt-sensi-63 de la Sierra A, del Mar Lluch M, Coca A, et al Assessment of salt-sensitivity in essential hypertension by 24-h ambulatory blood pressure monitoring Am J Hypertens 1995;8:970–977.

64 Wilson DK, Sica DA, Miller SB Effects of potassium on blood pressure in salt-sensitive and salt-resistant adolescents Hypertension 1999;34:181–186.

65 Weinberger MH, Miller JZ, Luft FC, et al Definitions and characteristics of sodium sensitivity and blood pressure resistance Hypertension 1986;8:II127–II134.

66 Falkner B, Kushner H, Khalsa DK, et al Sodium sensitivity, growth and family history of hypertension in young blacks J Hypertens 1986;4(Suppl 5): S381-S383.

67 Harshfield GA, Alpert BS, Pulliam DA, et al Electrolyte excretion and racial differences in nocturnal blood pressure Hypertension 1991;18:813–818.

68 Profant J, Dimsdale JE Race and diurnal blood pressure patterns Hypertension 1999;33: 1099–1104.

69 Yo Y, Nagano M, Moriguchi A, Nakamura F, Kobayashi R, Okuda N, et al Predominance of nocturnal sympathetic nervous activity in salt-sensitive normotensive subjects Am J Hyper- tens 1996;9:726–731.

70 Campese VM, Romoff MS, Levitan D, et al Abnormal relationship between Na + intake and sympathetic nervous activity in salt-sensitive patients with essential hypertension Kidney Int 1982;21:371–378.

71 Harshfield GA, Pulliam DA, Alpert BS Patterns of sodium excretion during sympathetic nervous system arousal Hypertension 1991;17:1156–1160.

72 Light KC, Koepke JP, Obrist PA, et al Psychological stress induces sodium and fluid retention

in men at high risk for hypertension Science 1983;229:429–431.

73 Grignolo AJ, Koepke JP, Obrist PA Renal function, heart rate, and blood pressure during exercise and avoidance in dogs Am J Physiol 1982;2:R482–R490.

74 Krishna GG, Miller E, Kapoor S Increased blood pressure during potassium depletion in normotensive man N Engl J Med 1989;320:1177–1182.

75 Weinberger MH, Luft FC, Block R, Henry DP, Pratt JH, et al The blood pressure-raising effects of high dietary sodium intake: racial differences in the role of potassium J Am Coll Nutr 1982;1:139–148.

76 Linas SL The role of potassium in the pathogenesis and treatment of hypertension Kidney Int 1991;39:771–786.

77 Fujita T, Ito Y Salt loads attenuate potassium-induced vasodilation of forearm vasculature in humans Hypertension 1993;21:772–778.

From: Contemporary Cardiology:

Blood Pressure Monitoring in Cardiovascular Medicine and Therapeutics

Edited by: W B White © Humana Press Inc., Totowa, NJ

Prognostic Value of Ambulatory

Blood Pressure Monitoring

PROGNOSTIC VALUE OF AMBULATORY BLOOD PRESSURE

AMBULATORY BLOOD PRESSURE AS A CONTINUOUS VARIABLE

OBSERVED VERSUS PREDICTED AMBULATORY BLOOD PRESSURE

WHITE COAT HYPERTENSION

WHITE COAT EFFECT

DAY–NIGHT BLOOD PRESSURE CHANGES

ULTRADIAN BLOOD PRESSURE VARIABILITY

AMBULATORY HEART RATE

AMBULATORY PULSE PRESSURE

is the high variability of BP, with consequent need of several measurements inorder to provide a more representative estimate of the average BP load to which

a given person is exposed over a given time interval On this assumption, cally hypertensive subjects may benefit from ambulatory BP monitoring (ABPM),

clini-a procedure which gclini-athers severclini-al BP meclini-asurements over time, when clini-a mclini-aincondition is fulfilled:

The difference in cardiovascular disease risk among different categoriesgenerated by ABPM should be greater than the difference among risk cate-gories generated by the standard measurement of BP As a result, statisticalmodels including different measures of ABPM should lead, after adjust-ment for concomitant risk factors, to a lesser unexplained risk of cardiovas-cular disease than models including different measures of clinic BP

In other words, ABPM should allow a clinically superior stratification of vascular risk when compared with the standard sphygmomanometric measure-ments of BP Once this point has been fulfilled on the basis of the results of onesingle session of the procedure, the next logical step would be the assessment ofthe prognostic value of serial changes of that procedure over time An associationbetween serial changes of the procedure and future occurrence of disease wouldmake the procedure not only a predictor but also a surrogate measure of disease.Once these aspects have been clarified, the procedure may be ready for inter-vention trials designed to see whether therapeutic interventions targeted on dif-ferent risk categories generated by the new procedure are more effective in terms

cardio-of disease reduction and/or cost-effectiveness than interventions on categoriesgenerated by the existing procedures

There is evidence from clinical studies published over the last few years anddiscussed in this review that one single session of ABPM provides clinically use-ful information that improves cardiovascular risk stratification based on clinic

BP and other traditional risk factors

PROGNOSTIC VALUE OF AMBULATORY BLOOD PRESSURE

The available observational studies on the prognostic value of ambulatory BPhave been conducted in tertiary care centers on subjects with essential hyper-

tension either untreated (1–15) or treated (16) at the time of execution of ABPM

or in the general population (17–20) (Table 1) In these studies, cardiovascular

morbidity and mortality were the main outcome measures and one single session

of ABPM allowed the definition of risk groups, which differed in their long-termoutcome even after adjustment for several potential confounders

Table 2 provides an overview of the results obtained by several independentresearch groups Because some groups produced more that one report on parti-ally overlapped populations, each group is represented in Table 2 with its largestcontribution in terms of patient-years of observation

We are involved in the PIUMA study (Progetto Ipertensione Umbria aggio Ambulatoriale) for 12 yr, an ongoing observational registry of morbidityand mortality in white adult subjects with essential hypertension The study pro-

Monitor-Table 1 Setting of Prognostic Studies with Ambulatory Blood Pressure Monitoring

Referred untreated subjects with essential hypertension (1–15)

Referred treated subjects with resistant hypertension (16)

General population (17–20)

Table 2 Observational Prognostic Studies with Ambulatory Blood Pressure Monitoring:

Contributions by Different Centers

No of Kind of Follow-up Total Fatal Author (ref.) Country Year subjects population (yr) events events

Full papers

Ohkubo et al (17) Japan 1997 1542 GP, U, T 5.1 n.r 93

Yamamoto et al (15) Japan 1998 105 RPS, T, U 3.2 15 n.r.

Verdecchia et al (8) Italy 1998 2010 RPH, U 3.8 200 36

Note: RPH = Referred patients with hypertension; GP = general population; RPS = referred

patients with stroke; U = untreated, T = treated, n.r = not reported.

tocol has been reported in detail (3–10) Follow-up is performed by family doctors

in cooperation with our outpatient clinic, and subjects are treated with the goal

of reducing clinic BP below 140/90 mmHg using standard lifestyle and cological measures There are periodical contacts with family doctors and tele-phone interviews with enrolled subjects in order to ascertain the vital status andthe occurrence of major cardiovascular complications

pharma-Contrary to the generally perceived opinion of paucity of observational nostic studies on ABPM, Table 2 reveals an unsuspected high number of exam-ined subjects (more than 6000) and subsequent outcome events (more than 500,

prog-at least 245 of which are fprog-atal) across different studies

Moreover, the survival analyses of other large databases, including the Cornell

Study (Pickering, personal communication), the OvA study (21), and the ELSA study (22), are expected in a near future For now, the existing database allows

one to consider the prognostic value of ABPM according to eight differentapproaches to data analysis (Table 3)

AMBULATORY BLOOD PRESSURE

AS A CONTINUOUS VARIABLE

An assessment of the prognostic value of ambulatory BP considered as a

con-tinuous variable has been carried out in the setting of the Ohasama study (18–20),

which is a general-population study in ambulant subjects aged 20 yr or moreliving in a Japanese rural area Some of the subjects were untreated and somewere being treated at the time of ABPM During follow-up, which lasted anaverage of 5 yr, there were 93 fatal cardiovascular events After adjustment forage, sex, smoking status, clinic BP, and use of antihypertensive medications, therisk of cardiovascular mortality was significantly increased in the highest quintile

of the distribution of average 24-h systolic BP, whereas no independent relationwas found between clinic BP and mortality There was a U-shaped relationshipbetween cardiovascular mortality and average 24-h systolic and diastolic BP,which may be interpreted as a possible expression of the link between low BPlevels and various morbid conditions in the general population

A limit of this study, which was the first to address the prognostic value ofABPM in the general population, is the lack of statistical adjustment for thepotential confounding effect of diabetes, serum cholesterol, and a family history

of premature coronary heart disease, three potent prognostic predictors

Another relevant study is that by Redon et al (16) In this study, 86 patients

with diastolic BP > 100 mmHg despite treatment with three or more drugs, ing a diuretic, underwent 24-h ABPM During an average follow-up period of

includ-4 yr, 21 patients developed a first cardiovascular morbid event After adjustmentfor age, sex, smoking, left ventricular hypertrophy, and clinic BP, the event rate

was higher ( p < 0.02) in the upper (13.6 events per 100 patient-years) than in the

middle (9.5 events per 100 years) and lowest (2.2 events per 100 years) tertile of daytime diastolic BP Despite some limitations, including the

patient-Table 3 Prognostic Studies with Ambulatory Blood Pressure Monitoring:

Different Analytical Approaches

Ambulatory BP as a continuous variable (16.18.19) Observed versus predicted ambulatory BP (1,2,23)

“White coat” hypertension (3,11,12,24)

“White coat” effect (6) Day–night BP changes (3,4,10,13–15,19,25) Ultradian BP variability (26)

Ambulatory heart rate (7) Ambulatory pulse pressure (8)

small sample size and the lack of statistical adjustment for the potential founding effect of factors like serum cholesterol and family history of prematurecoronary heart disease, this study is the first to support an independent predictivevalue of ABPM in patients with resistant hypertension.

con-Ambulatory BP has been examined as a continuous variable in the Systolic

Hypertension in Europe (Syst-Eur) study (26) In that study, ambulatory BP

monitoring was carried out at randomization in 808 untreated patients Of these,

98 developed a cardiovascular event over the follow-up period After statisticaladjustment for age, sex, office BP, active treatment, previous events, cigaretsmoking, and residence in western Europe, nighttime systolic BP was an indepen-dent predictor of total, cardiac, and cerebrovascular events, whereas the averagedaytime BP did not achieve significance (Fig 1) In the subjects randomized

to placebo, for every 10% higher night/day ratio of systolic BP, the risk of events

increased by 41% (95% confidence cardiovascular intervals: 3–94%; p = 0.03).

These findings strongly support the prognostic value of ambulatory BP, in ticular for BP levels recorded during the night

par-OBSERVED VERSUS PREDICTED AMBULATORY BLOOD PRESSURE

If we plot (Fig 2) clinic BP vis-à-vis the average daytime ambulatory BP in

a large population of subjects with essential hypertension, it is apparent that forany given value of clinic BP, the observed ABP is seldom that predicted by alinear regression equation, whereas it is often considerably higher or lower than

Fig 1 Relative hazard rates for total cardiovascular events for every 10 mmHg increase

in systolic blood pressure after adjustment for age, sex, smoking, office systolic blood pressure, previous cardiovascular events, and residence in western Europe.

predicted For example, in patients with clinic systolic BP of 140–150 mmHg,the average daytime ABP may swing between less than 100 to about 190 mmHg.Following this analytical approach, the ambulatory BP averages are not con-sidered in absolute terms, but in relation to the values of clinic BP in that particu-lar subject Hence, a given ABP average may be that predicted by the regressionequation if it coincides with the regression line, or it may be lower or higher thanpredicted Dorothee Perloff, Maurice Sokolow, and colleagues, the pioneers ofclinical use of ABPM, were the first to note that for any given value of clinic BP,the target organ damage in hypertension was more consistent in the patients with

higher-than-predicted ABP than in those with lower-than-predicted ABP (1,2).

Subsequently, they followed for an average of 5 yr, 1076 patients with essentialhypertension and detected 153 cardiovascular morbid events, 75 of which werefatal The risk of events was significantly higher in the subset with higher-than-predicted ABP than in that with lower-than-predicted ABP, particularly in sub-jects with stage I hypertension This study is limited by the lack of a normotensivecontrol group, the lack of nocturnal blood pressure monitoring (resulting fromthe use of manually activated recorders), the lack of serum cholesterol, and cigaretsmoking among the potential confounders in the multivariate analysis Despite

Fig 2 The plot shows the association between clinic blood pressure and average daytime

ambulatory blood pressure in 2010 untreated subjects with essential hypertension (PIUMA database).

these limitations, this landmark study clearly showed for the first time the mous potential clinical value of noninvasive ABPM and opened the way toward

enor-a lenor-arger use of this dienor-agnostic technology in the clinicenor-al setting

It is important to clarify that the subjects with lower-than-predicted ABP donot have a “normal” ABP Hence, their cardiovascular risk should not be consid-ered analogous to that of clinically normotensive subjects In a word, the twoconcepts of “lower-than-predicted ABP” and “white coat” hypertension (dis-cussed next) must be kept separate On the other hand, the subjects with higher-than-predicted ABP are clearly characterized by office underestimation of theusual levels of BP This high-risk group would remain undiagnosed with the soleuse of clinic BP

Some years ago, we have shown (27) that cigaret smoking is an important

determinant of a higher-than-predicted ambulatory BP In fact, clinic BP is lesslikely to be affected by smoking (usually, patients quit smoking shortly beforethe clinical visit), contrary to what happens to ambulatory BP, and the effect of

smoking on ambulatory BP may lead to left ventricular (LV) hypertrophy (27).

To clarify the prognostic impact of this phenomenon, we have recently analyzedthe outcome of 841 subjects with essential hypertension JNC VI stage I who werefollowed for about 4 yr During this period, the rate of cardiovascular morbidevents was about 1 per 100 patient-years in the lowest quintile versus about 3per 100 patient-years in the highest quintile of the difference between observed

and predicted ABP (log-rank test: p = 0.012) However, such a difference did

not remain significant in a multivariate survival analysis, which included cigaretsmoking, age, 24-h pulse pressure, white coat hypertension, and a nondipping

pattern as independent prognostic determinants (23).

Taken together, all these data suggest that a higher-than-predicted ABP should

be considered a univariate prognostic predictor in subjects with stage I sion Its adverse impact, however, is less strong than that of aging, cigaret smok-ing, and other more predictive components of ambulatory BP

hyperten-WHITE COAT HYPERTENSION

White coat hypertension, also referred to as office hypertension or isolatedclinic hypertension, is generally defined by a persistently elevated office BPtogether with a normal pressure outside the office Although the usual definition

of elevated office BP is out of discussion (140 mmHg systolic and/or 90 mmHg

diastolic) (28,29), there is great deal of controversy about the definition of

nor-mal BP outside the office As shown in Table 4, it is hard to find two studies usingthe same definition of white coat hypertension based on results of ABPM Somestudies use lower cutoff points, whereas others use higher cutoff points; thedefinition is based on systolic values in some studies and on diastolic values inother studies Some studies use the average BP during daytime and other studies

use the average 24-h BP Still, some studies include a measure of the office–ambulatory BP difference in the definition.

At a first glance, the differences between the upper normal limits of ABP used

to define white coat hypertension might seem small and of little clinical

rele-vance However, we have shown (32) that not only the prevalence of white coat

hypertension but also LV mass at echocardiography and the prevalence of LVhypertrophy increased markedly when swinging from more restrictive (lower)

to more liberal (higher) limits of ambulatory BP normalcy used for the definition

of white coat hypertension (Fig 3)

Inspection of Fig 4, drawn from the PIUMA dataset (49), indicates the

impor-tance of a restrictive definition of the upper normal limits of ambulatory BP inorder to identify a population with characteristics of potentially low cardiovas-cular risk The figure shows that the prevalence of LV hypertrophy, virtuallyabsent below 120 mmHg and very modest below 130 mmHg (6%), increased to10.5% when the limit was set to 140 mmHg Thus, even modest swings over arelatively narrow range of presumably normal or nearly normal ambulatory BPresult in considerable differences in the prevalence of subjects with increased

LV mass and, because of its adverse prognostic value (3,50,53), with potentially

increased cardiovascular risk

White et al used a restrictive definition of white coat hypertension (i.e., age daytime ambulatory BP < 130/80 mmHg) and found normal values of LV

aver-Table 4 Different Definitions of White Coat Hypertension Based on ABPM

Pickering et al (30) < 134/90 mmHg (daytime ambulatory BP)

White et al (31) < 130/80 mmHg (daytime ambulatory BP)

Verdecchia et al (32) < 131/86 mmHg (W) or 136/87 mmHg (M) (daytime ambulatory BP)

Pierdomenico et al (33) < 135/85 mmHg (24-h ambulatory BP)

Kuwajima et al (34) < 140 mmHg (24-h ambulatory BP)

Cardillo et al (35) < 134/90 mmHg (daytime ambulatory BP)

Cerasola et al (36) < 134/90 mmHg (daytime ambulatory BP)

Glen et al (37) < 95 mmHg (daytime ambulatory BP)

Siegel et al (38) < 135/85 mmHg (daytime ambulatory BP)

Weber et al (39) < 85 mmHg (24-h ambulatory BP) and  5 mmHg < clinic BP

Hoegholm et al (40) < 90 mmHg (daytime ambulatory BP)

Marchesi et al (41) < 135/91 mmHg (daytime ambulatory BP)

Bidlingmeyer et al (42) < 140/90 mmHg (daytime ambulatory BP)

Rizzo et al (43) < 142/90 mmHg (daytime ambulatory BP)

Trenkwalder et al (44) < 146/87 mmHg (daytime ambulatory BP)

Staessen et al (45) < 133/82 mmHg (24-h ambulatory BP)

Amar et al (46) < 131/86 mmHg (W) or 136/87 mmHg (M) (daytime ambulatory BP)

Cuspidi et al (47) < 135/85 mmHg (24-h ambulatory BP)

Polonia et al (48) < 132/74 mmHg (daytime ambulatory BP)

W = women, M = men.